Abstract | ||
---|---|---|
We study a distributionally robust mean square error estimation problem over a nonconvex Wasserstein ambiguity set containing only normal distributions. We show that the optimal estimator and the least favorable distribution form a Nash equilibrium. Despite the non-convex nature of the ambiguity set, we prove that the estimation problem is equivalent to a tractable convex program. We further devise a Frank-Wolfe algorithm for this convex program whose direction-searching subproblem can be solved in a quasi-closed form. Using these ingredients, we introduce a distributionally robust Kalman filter that hedges against model risk. |
Year | Venue | Keywords |
---|---|---|
2018 | ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 31 (NIPS 2018) | nash equilibrium,convex program,estimation problem,frank-wolfe algorithm,model risk,normal distributions,wasserstein distributionally robust kalman filtering |
DocType | Volume | ISSN |
Conference | 31 | 1049-5258 |
Citations | PageRank | References |
1 | 0.35 | 11 |
Authors | ||
5 |
Name | Order | Citations | PageRank |
---|---|---|---|
Soroosh Shafieezadeh-Abadeh | 1 | 33 | 3.01 |
Viet Anh Nguyen | 2 | 127 | 19.08 |
Daniel Kuhn | 3 | 559 | 32.80 |
Peyman Mohajerin Esfahani | 4 | 1 | 1.37 |
Mohajerin Esfahani, Peyman M. | 5 | 1 | 0.35 |